As electronic devices become ever smaller, the importance of inductively- and capacitively-induced crosstalk or noise becomes more significant. This has proven especially true in respect of integrated circuit- or chip-based differential signal electronic circuits (where signals transmitted and received by one or more chips in such circuits are typically 180 degrees out of phase respecting one another).
The severity of inductively- and/or capacitively-induced crosstalk noise in differential signal circuits depends upon a number of different factors, including the number and type of physical interconnects in the circuit, the number and type of integrated circuits or chips incorporated into such circuits, the physical proximities of electrical conductors to electromagnetic noise sources, the numbers, configurations and types of electrical conductors employed in a given printed circuit board (PCB) design. If the impact of crosstalk noise is not taken into account when a differential signal electronic circuit is designed, the circuit may fail when employed in actual practice.
In view of the foregoing difficulties, it is desirable to measure the amount and character of inductively- and/or capacitively-induced crosstalk noise in differential signal electronic circuits under varying broadband noise coupling conditions.
Several different methods and devices are currently employed to effect such measurements, including: (1) power splitter test networks such as the Agilent Part No. 11667B; (2) pick-off T circuits such as those manufactured by Picosecond Pulse Labs in Boulder, Colo.; and (3) directional coupler circuits.
The foregoing methods and devices suffer from various shortcomings, however. While the Agilent Power Splitter Part No. 11667B may be employed to combine an aggressor signal with a victim signal, losses of 6 dB result. Additionally, Agilent Power Splitter Part No. 11667B is not tailored for use in differential signal circuits. Power splitter and resistor network circuits also consume significant amounts of power and cause victim and aggressor signal attenuation. Resistor networks require considerable aggressor signal attenuation so as to provide low loss victim paths.
Various embodiments of pick-off T circuits may suffer from mismatched impedances, variations in frequency response and unwanted reflections of aggressor signals.
Directional couplers can be very expensive; Agilent's Part No. 87301D costs around $1,900 each.
What is needed are methods and devices for the measurement of inductively- and/or capacitively-induced crosstalk noise in differential signal electronic circuits that are broadband, inexpensive, consume little power, and that do not significantly alter or affect the fidelity of the signals being measured and evaluated.